How Do You Measure SEO Performance When AI Answers Reduce Clicks?

For years, measuring SEO performance has been relatively straightforward.

Rankings improve. Impressions increase. More people click through to your website. Traffic grows, and ideally, so do conversions.

But AI is changing how people interact with search results.

As AI search optimisation becomes a growing consideration for businesses, the way we assess search visibility needs to evolve too.

Google AI Overviews and AI Mode can answer questions directly within search results. Platforms such as ChatGPT and Perplexity also help users research products, compare services and evaluate businesses without necessarily visiting the websites that informed their answers.

A business could be appearing in more relevant searches, being referenced in AI-generated answers and building awareness, while receiving fewer website visits.

Does that mean its SEO is performing worse?

Not necessarily.

The challenge is that traditional SEO reporting was largely designed around getting people from a search engine onto a website.

AI search introduces another possibility: people can discover and evaluate a business without making that click.

So how should we measure SEO performance when traffic no longer tells the full story?

SEO performance needs more than one measure of success

SEO should now be evaluated across three areas:

  • Search visibility: How often is your business appearing in relevant search results, including AI-generated experiences?
  • Website engagement: When people visit, are they engaging with relevant content and taking meaningful actions?
  • Business outcomes: Is organic search contributing to enquiries, sales and revenue?

These measures tell different parts of the story.

An increase in visibility doesn’t automatically mean more traffic. More traffic doesn’t necessarily mean more customers.

And fewer clicks don’t always mean SEO is delivering less value.

The objective is to understand what’s changing at each stage.

Start by separating search visibility from traffic

One of the first things to investigate when organic traffic declines is whether search visibility has declined alongside it.

Google Search Console remains an important starting point.

Its traditional Performance report provides four useful metrics:

  • Impressions
  • Clicks
  • Click-through rate (CTR)
  • Average position

Looking at these together is more valuable than analysing clicks alone.

Consider this hypothetical example of an informational article.

Metric Previous period Current period Change
Impressions 100,000 120,000 +20%
Clicks 5,000 3,600 -28%
CTR 5.0% 3.0% -40%
Average position 4.2 4.1 Slight improvement

If we looked at clicks alone, we’d conclude that SEO performance had declined significantly.

But impressions increased, and average position remained stable.

The website is still appearing in search results, yet fewer people are clicking through.

AI-generated answers could be contributing to this change. So could featured snippets, changes in search results or shifts in user behaviour.

The data doesn’t prove the cause, but it tells us where to investigate.

A decline in clicks is not necessarily the same as a decline in search visibility.

How do you measure visibility in AI search?

Bing was one of the first major search platforms to introduce dedicated AI visibility reporting. In February 2026, Bing Webmaster Tools launched its AI Performance report, showing how website content is cited across Microsoft Copilot, AI-generated Bing summaries and selected partner experiences.

Rather than just showing whether a page appeared, Bing reports total citations, the pages being cited, related grounding queries and how citation activity changes over time.

It has since expanded this with preview insights into topics, search intent and citation share, giving website owners more context about where their content is being referenced.

Google followed with its Generative AI Performance report in Search Console in June 2026, providing a dedicated view of impressions from AI Overviews and AI Mode. Businesses can see which pages appear in these experiences and analyse trends by country, device and date.

Bing focuses more on citations and the queries associated with them, while Google’s dedicated reporting focuses on AI search impressions.

For example, an informational article might receive fewer traditional clicks while continuing to gain exposure through Google AI impressions or citations in Bing’s AI experiences.

But there’s an important limitation. Impressions and citations indicate exposure; neither tells us whether someone engaged with the business or became a customer.

Neither platform offers a complete view of AI visibility across every service, including ChatGPT and Perplexity.

Third-party AI visibility tools can help investigate brand mentions, competitor recommendations and source citations across a selected set of prompts.

Understanding how AI platforms select and cite sources can also provide context for interpreting brand mentions and the sources behind AI-generated answers.

But these results are generally based on sampled AI responses, not a complete record of real customer interactions.

They’re useful for understanding patterns, identifying gaps and monitoring changes, rather than treating every visibility score as an exact measure of market share.

Not all lost clicks have the same business impact

Someone searching for “what is a heat pump dryer” is at a different stage of the buying journey from someone searching for “buy 8kg heat pump dryer Australia“.

The first person may simply want an explanation.

The second may be ready to purchase.

AI answers can influence both searches, but losing a click from each has potentially different commercial consequences.

That’s why we recommend segmenting SEO performance by search intent.

Search intent Example What matters
Informational What is a heat pump dryer? Visibility, AI exposure, engagement
Commercial Best heat pump dryers for apartments Product discovery, engagement, conversions
Transactional Buy 8kg heat pump dryer Purchases, revenue, conversion rate

Imagine informational blog traffic declines by 25%, while commercially important category and product pages maintain traffic and generate more revenue.

That’s different from losing 25% of visits to your highest-converting product categories.

Sitewide traffic figures can hide these differences.

The more useful question is:

Where are we losing traffic, what was that traffic worth, and is the business impact proportionate to the decline?

Measure what happens after visitors arrive

Even when search traffic declines, the visitors who do arrive may be more qualified.

Someone who has already received a general explanation from an AI answer might visit a website when they’re ready to compare products, request a quote or make a purchase.

This is where GA4 becomes important.

For a closer look at these visits, see our guide to tracking AI referral traffic in GA4, including traffic from platforms such as ChatGPT and Perplexity.

Rather than stopping at organic sessions, look at measures such as engagement, product views, retailer interactions, form submissions, purchases and revenue.

For example:

Metric Previous period Current period Change
Organic sessions 50,000 40,000 -20%
Purchases 500 550 +10%
Conversion rate 1.0% 1.38% Improved
Organic revenue $75,000 $88,000 +17.3%

These are illustrative figures, but they highlight an important point.

Organic traffic declined by 20%, yet purchases and revenue increased.

That doesn’t mean the business should ignore the lost traffic. There may still be opportunities to recover valuable visits.

But describing SEO performance as being down 20% would be misleading.

The next step is understanding why fewer visits are producing better commercial outcomes.

Are lower-intent informational visits declining? Has the landing-page mix changed? Have conversion rates improved?

Those answers are more useful than the overall traffic figure.

What should businesses be measuring as AI search evolves?

As AI search continues to develop, businesses may need to rethink what they consider successful SEO performance.

Traditional metrics such as rankings, impressions and organic traffic remain important. But they don’t always capture the full picture of how people discover and engage with a brand.

A more complete approach to SEO measurement could consider five areas:

Measurement area Key metrics
Search visibility GSC impressions, rankings, non-brand visibility
AI visibility Google AI impressions, monitored brand mentions and citations
Website traffic Organic sessions, clicks, CTR
Engagement Engaged sessions, product views, meaningful interactions
Business outcomes Leads, purchases, conversion rate, revenue

Not every business needs to track every metric immediately. AI visibility measurement is still evolving, and the right approach will depend on the business, its objectives and the data available.

What’s more important is understanding the limitations of looking at organic traffic in isolation.

For example, an ecommerce business may benefit from separating product pages, category pages and informational guides when evaluating performance.

A service-based business might instead focus on service pages, industry pages and educational content.

As we explored in our guide to diagnosing enterprise SEO problems at scale, sitewide averages can hide what’s happening in the parts of a website that matter most.

The same principle applies as AI becomes a larger part of how people search.

An effective enterprise SEO strategy therefore needs to connect these performance signals to commercial priorities, rather than relying on rankings and traffic alone.

The goal isn’t to add more metrics for the sake of reporting. It’s to understand which measures actually help explain SEO’s contribution to the business.

SEO performance hasn’t stopped being measurable

AI search is changing the relationship between search visibility, website visits and business outcomes.

A business might gain AI exposure while receiving fewer clicks. Another might lose informational traffic while increasing conversions and revenue.

Neither situation can be evaluated accurately using traffic alone.

The better approach is to connect traditional search performance with AI visibility, engagement and commercial results.

That means understanding where your business appears, which pages attract visitors, what those visitors do and whether SEO is contributing to meaningful business outcomes.

SEO performance hasn’t stopped being measurable. We just need to be more careful about what we’re measuring.

At Intender, we help businesses move beyond surface-level SEO reporting to understand what’s driving organic performance, where opportunities are being lost and which improvements deserve priority.

Explore our SEO and AI discovery services to learn how we approach search strategy and performance measurement.

FAQs

Can SEO be successful if organic traffic declines?

Yes. Traffic can decline while conversions and revenue remain stable or increase. However, businesses should investigate whether valuable search visibility or customer acquisition opportunities are being lost.

Does Google Search Console track AI Overviews?

Yes. Google includes AI Overview and AI Mode activity within its overall Search performance reporting. Its dedicated Generative AI Performance report also provides impression data for supported AI search experiences.

Can GA4 track traffic from ChatGPT?

GA4 can identify some visits referred from ChatGPT and other AI platforms when referral information is available. However, it cannot measure people who see a brand mentioned in an AI answer but never visit the website.

LinkedIn
WhatsApp
Reddit
X

Your ads are reaching millions. Converting? 

Not so much.

We target people already showing buying signals 
— not just browsing. We call them intenders.

Stop paying for attention. Start getting intent.